A common feature of the developing nervous system is that many immature neural circuits generate spontaneous activity. This phenomenon has been studied most extensively in the developing visual system. Before vision is possible in the retina, neighboring retinal ganglion cells, the retina's projection neurons, spontaneously fire correlated bursts of action potentials separated by long periods of silence. This bursting activity propagates across the retina in the form of "retinal waves". Retinal waves are required for the proper refinement of retinal projections to its central target, the visual thalamus and the superior colliculus. How are retinal waves generated? We propose to use a combination of transgenic mice and whole cell patch clamp recordings from acutely isolated mouse retina to elucidate the synaptic basis of retinal waves. Specifically, we will determine the relative role of excitatory synaptic inputs, inhibitory synaptic inputs and gap junctions. These experiments will help us to better understand the mechanisms by which cells in the brain form the correct connections during development. The knowledge gained from these experiments may help to explain congenital defects of the visual system and could be used to help reestablish connections in the injured brain.